MiR-27b-3p targeting BDNF inhibits TrkB / CREB signaling pathway and improves IL-1 β induced chondrocytic inflammation

improves IL-1 β induced chondrocytic inflammation Cailian Ruan1,Rui junCong2,MiaoWang2,LiJunWang3,YongYu4, XiaoJiLi#1,HaiXiaLv#5 1College of Medicine, Yan'an University, Yan'an, 716000,Shaanxi,China. 2Tenth People's Hospital Affiliated to Tongji Universityi,200092,Shanghai,China. 3Shaanxi Normal University,Xi'an,710061,Shaanxi ,China. 4Shaanxi Geriatric Hospital,Xi'an,710061,Shaanxi ,China. 5Xi'an Jiaotong University,Xi'an,710061,Shaanxi ,China.


Introduction
Meniscus is the fibrous cartilage plate in the knee joint,the inside is a "C"shape and the outside is an "O"shape.The outer edge of the meniscus is thick, the inner edge is thin and concave, which is composed of collagen fibers, is an important part of the structure and function of the knee joint.It has the properties of relieving vibration, lubricating joints, reducing friction, increasing stress area, conducting load and maintaining the stability of knee joint (1).Trauma or natural aging can make meniscus appear different degrees of deformation, and then cause surrounding tissue edema、 proliferation、make the knee joint lose normal function and stability, accelerate the degeneration of articular cartilage, and further lead to the occurrence of Knee Osteoarthritis (KOA).
Knee Osteoarthritis (KOA) is a widely spread disease that characterized by articular cartilage degradation,subchondral bone sclerosis,hyperostosis, osteochondral vessel formation and articular inflammatory reaction. KOA is a chronic progressive disease and a leading cause of disability (2). Patients with KOA suffer from joint pain, swelling and decreased living quality (3). More than 10% of adults over 60 have KOA, and the incidence is increasing with age and Body Mass Index (BMI) (4). However, there are no effective treatment therapies to reverse this process of KOA in clinic.
Thus,clarifying the mechanism of KOA is critical to develop effective methods to treat KOA.
MicroRNAs are a cluster of non-coding RNAs with 18-15 neucleotides in length.
They could be bind to the 3' untranslated region of the target gene mRNA (5,6). This bind could induce the degradation of mRNA or inhibit the translation of target gene mRNA (7,8). MiRNAs have widely proved to regulate cellular behavior in multi diseases, like cancer, spinal cord injury, hepatitis etc (7,9). MiRNAs could regulate the process of KOA. MiR-140 is expressed in chondrocytes and regulate cartilage development (10).MiR-26a could regulate KOA progression via targeting FUT4/NF-κB pathway (11). MiR-489-3p has been reported to regulate tumorigenesis, bladder cancer cell migration and proliferation and prostate cancer progression (12)(13)(14). However, the role of miR-27b-3p in KOA has not been clarified.
The level of IL-1β is increasing in the cartilage tissue in KOA patients (15,16).
IL-1β is regarded as one of the two inflammatory factors in KOA onset and development (15). The increasing of other cytokines, like IL-6, could be induced by IL-1β. The catabolism is also induced by IL-1β, which results in cartilage degradation.
Therefore, inhibiting the role of IL-1β is critical to reduce the process of KOA (17).
Chondrocytes is the main cell type in cartilage. The main function of chondrocytes is to maintain the metabolism balance of cartilage.They can secrete collagen and proteoglycan to maintain the integrity and function of cartilage. However, IL-1β could also induce chondrocytes apoptosis to decrease the self-repair ability of cartilage (18).
Thus, inhibiting chondrocyte apoptosis is of great meaning to treat KOA.
TrkB/CREB pathway is important in many bioprocess, like inflammation response, cell injury response and apoptosis (19). In KOA, the TrkB/CREB pathway is activated by IL-1β (20). Inhibiting activation of TrkB/CREB could decrease IL-1β induced chondrocyte apoptosis and showed beneficial effect on KOA (20). Evidence has showed brain-derived neurotrophic factor (BDNF) is upregulated in the joint tissues of KOA patients (21,22). BDNF is an extracellular cytokine that could actively TrkB/CREB pathway by binding to its cell membrane receptor, NTR. Thus, inhibiting BDNF could be an efficient way to inhibit the development of KOA via reducing TrkB/CREB activation.
In this study, we aimed to clarify the role of miR-489-3p in KOA in both in vitro and in vivo. The target gene of miR-27b-3p was confirmed to be BDNF and miR-27b-3p could regulate chondrocyte apoptosis via BNDF/TrkB/CREB pathway.
This study could provide novel treatment target for KOA.

Animal and Ethic statement
A total of 96 male SD rats(250±0.05)g were obtained from Yan'an University Laboratory Animal Technology Co., Ltd. The rats were randomly separated into four groups: Control, KOA, KOA+Len-NC and KOA+Len-miR-27b-3p groups. Each group had 24 rats for RT-qPCR, histological detection, western blotting and ELISA.
The rats were housed in 12 h light/dark cycles. All the procedures were approved by the ethics committee of Yan'an University.

Rat KOA Model
To detect the role of miR-27b-3p in KOA, rat KOA model was established as previously reported . The KOA rat model was established by transection of lateral meniscus of right posterior limb lateral meniscus cutting off.Rats were anesthetized by intraperitoneal injection of pentobarbital sodium, the skin of the right posterior limb lateral meniscus cut off. A para patellar incision was made, the patella was dislocated. Then, the lateral meniscus was cut. Then, the incision was sutured. The lentivirus overexpressing miR-27b-3p (1×10 9 units in 20 μL, twice a week for 7 weeks, n=24) was injected into the space of keen joint. After 8 weeks, rats were sacrificed through carbon dioxide and joint tissue samples were collected.
The cell culture medium was changed every day. IL-1β (10 ng/mL, Sigma, USA) was applied to induce KOA in vitro. BDNF (10 ng/mL, Sigma, USA) was applied to treat cultured chondrocytes.

RT-qPCR
To detect the expression of BDNF, iNOS, COX-2 mRNA and miR-27b-3p in cultured chondrocytes and the expression of miR-27b-3p in cartilage tissue, the RT-qPCR was conducted as previously reported. The cell samples and cartilage tissues were harvested and the total RNA was extracted with TRIzol kit (Invitrogen, USA). Nanodrop 2000 was applied to determine the RNA concentration of samples.
The cDNA of miR-489-3p was synthesized with ReverAid First Strand cDNA Synthesis (Thermo, USA). The cDNA of BDNF, iNOS and COX-2 mRNA was obtained with Primescript RT Master Mix (Takara, Japan). MiScript SYBR® Green PCR Kit (Thermo, USA) was applied to conduct RT-qPCR assay. GAPDH mRNA was applied as internal reference of BDNF, iNOS and COX-2 mRNA, while U6 was set as internal reference of miR-27b-3p. The relative expression of these genes was calculated with 2 -ΔΔCt method. The primers were shown in Table 1.

Western Blotting
The protein expression of caspase-3 、 caspase-9 、 iNOS 、 COX-2 、 BDNF 、 TrkB,CREB、p-CREB in cultured chondrocytes and the protein expression of BDNF、 p-TrkB 、 p-CREB in cartilage tissues was detected with Western blotting. The total proteins in cells and tissues were harvested with RIPA buffer (Beyotime, China).
Nanodrop 2000 was applied to determine the protein concentration of samples. Then, proteins in samples were separated by electrophoresis in SDS-PAGE. Then, the proteins were transferred onto PVDF membranes. The PVDF membranes were blocked with non-fat milk (5%). Then, the primary antibodies were applied to label target proteins. Then, the secondary antibody (goat-rabbit IgG-HRP 1:1000) was applied to incubate PVDF membranes. Then, ELC kit (Beyotime, China) was applied to show the blots. GAPDH was set as an internal reference. Antibodies were purchased from Abcam, UK.

ELISA Assay
The concentration of nitric oxide, PGE2, TNF-α and IL-6 in cultured chondrocytes, and the concentration of Bax, Bcl-2, TNF-α and IL-6 in cartilage tissues were detected with ELISA assay. Briefly, the samples were prepared according to the manufacturer's instruction. The ELISA kits were purchased from R&D, USA.

Oligonucleotide Transfection
To detect the role of miR-27b-3p in KOA, the miR-27b-3p mimics and NC were transfected into cultured chondrocytes with the help of Lipofectamine2000 (Invitrogen, USA). These oligonucleotides all obtained from GenePharma, China. The cultured chondrocytes were seeded into culture plate at a density of 1×10 6 / mL. The transfection was conducted when the confluence reached 80%. 24 h after transfection, the cultured medium was replaced with fresh culture medium.

Flow Cytometry Assay
To demonstrate the role of miR-27b-3p on chondrocyte apoptosis, flow cytometry assay was conducted with FITC-AnnexinV apoptosis detection kit The acquired data were analyzed with FlowJo software.

Dual-luciferase reporter assay
To clarify whether BDNF is the target gene of miR-27b-3p, dual-luciferase reporter assay was conducted. The binding sites of miR-27b-3p on BDNF mRNA were predicted by TargetScan database. The BDNF mRNA 3'UTR was cloned into p-MIR-GLO reporter assay (Promega, USA). The reconstructed plasmid was named BDNF-WT. In the same way, the BDNF-MUT plasmid with mutant binding sites was reconstructed. The cultured chondrocytes were co-transfected with miR-27b-3p mimics/NC and BDNF-WT/BDNF/MUT with lipofectamine2000. 48 h after transfection, the relative luciferase activities were detected by dual-luciferase reporter assay system (Promega, USA).

Histological analysis
The cartilage tissues were harvested and fixed with paraformaldehyde (4%). The fixed tissues were embedded into paraffin and sliced (5 μm). The cartilage tissues were stained with safranin O staining. The pictures were obtained with the microscope (Nikon, Japan).

Statistical analysis
All data were recorded and analyzed with GraphPad Prism 6.0. The recorded data were presented as mean ± SD. The differences between more than three groups were analyzed by one-way ANOVA, followed by Turkey's multiple comparison test.
P<0.05 was considered as statistically significant.

MiR-27b-3p inhibition of IL-1β induced chondrocyte inflammation CH8 chondrocyte apoptosis
To further clarify the role of miR-27b-3p in KOA, our experiment transfected IL-1β treated chondrocytes with mir-27b-3p mimics.The expression of miR-27b-3p was greatly inhibited in IL-1β group compared with that in control group. This inhibition effect of IL-1β was totally reversed after miR-27b-3p mimics transfection ( Fig.1A). IL-1β-treated chondrocytes can cause chondrocyte inflammation, the possible mechanism is that IL-1β can induce chondrocyte apoptosis.while miR-27b-3p mimics reduced the percentage of apoptotic cells (Fig.1B). IL-1β also induced the expression of caspase-3 and caspase-9, while miR-27b-3p greatly inhibited the role of IL-1β (Fig.1C). These results indicated that miR-27b-3p has a protective effect on chondrocyte apoptosis induced by IL-1β.

MiR-27b-3p can reduce the expression of inflammatory factors in IL-1β treated chondrocytes
IL-1β induced chondrocytes have inflammatory reaction and apoptosis, is there a certain relationship between them? Do miR-27b-3p play an anti-apoptotic role by inhibiting inflammatory response?To verify this assumption,we detected the expression of iNOS and COX-2 mRNA expression with RT-qPCR. IL-1β could be greatly induced iNOS and COX-2 mRNA expression, while miR-27b-3p evidently reduced their expression ( Fig.2A). Similarly, miR-27b-3p could also reduce IL-1β upregulated iNOS and COX-2 protein expression (Fig.2B). The expression of nitric oxide, PGE2, TNF-α and IL-6 was also promoted by IL-1β treatment, and this promotion was greatly inhibited by miR-27b-3p mimics transfection (Fig.2C-F).

Bioinformatics analysis of BDNF is the target relationship of miR-27b-3p
To further clarify the mechanism of action of MIR-27b-3p in KOA,the target gene of miR-27b-3p was predicted in Targetscan database, and BDNF was predicted to be the target gene of miR-27b-3p. The predicted binding sites were showed (Fig.3A). Dual-luciferase reporter assay showed the relative luciferase activity was greatly reduced in BDNF-WT+miR-27b-3p mimics group compared with the other three groups (Fig.3B).Then, the expression of BDNF mRNA was detected after miR-27b-3p mimics treatment. IL-1β greatly upregulated the expression of BDNF mRNA, while miR-27b-3p evidently inhibited the role of IL-1β on BDNF mRNA expression (Fig.3C). Similarly, the expression of BDNF protein was also promoted by IL-1β and inhibited by miR-27b-3p (Fig.3D). These results indicated BDNF was the target gene of miR-27b-3p .

MiR-27b-3p targeted BDNF regulate downstream TrkB/CREB pathways
To further clarify the mechanism of miR-27b-3p on KOA, we predicted the downstream pathway of miR-27b-3p/BDNF. Previous studies have shown that TrkB/CREB is the downstream pathway of miR-27b-3p /BDNF (22). This study further verifies this conclusion,we detected the expression of TrkB、p-TrkB、CREB and p-CREB in cultured chondrocytes. The expression of p-TrkB and p-CREB was greatly induced after IL-1β treatment, while their expression was inhibited by miR-27b-3p mimics (Fig.4 A). These results indicated miR-27b-3p could inhibit the TrkB/CREB pathway that triggered by IL-1β. To further confirm the regulative role of miR-27b-3p on BDNF triggered TrkB/CREB pathway, BDNF was applied to cultured chondrocytes. Unregulated expression of p-TrkB and p-CREB that induced by IL-1β was further increased after adding BDNF. Unregulated expression of p-TrkB and p-CREB that performed by IL-1β+BDNF was inhibited by miR-27b-3p mimics treatment (Fig.4 B). These results indicated miR-27b-3p could target BDNF to regulate TrkB/CREB pathway.

In vivo experiments have shown that MiR-27b-3p reduces the degradation of extracellular matrix by inhibiting the expression of inflammatory factors, protects cartilage tissue, and relieves KOA pain
The above in vitro experiments prove that MiR-27b-3p can reduce KOA pain by inhibiting chondrocyte apoptosis and inflammation. In order to elaborate on this function of MiR-27b-3p, we further verify it through in vivo experiments. We established the rat model of KOA by knee meniscus dissection and injected miR-27b-3p mimics into the knee lumen to overexpress miR-27b-3p in the cartilage.
In KOA model group, the expression of miR-27b-3p was evidently inhibited compared with that in control group. The application of Len-miR-27b-3p evidently increased the expression of miR-27b-3p compared with that in Len-NC group (Fig.6A). These results indicated miR-27b-3p was decreased in KOA cartilage tissues and Len-miR-27b-3p could increase miR-27b-3p expression of cartilage tissues. Then, the cartilage tissues were harvested and stained with Safranin O/Fast green. The cartilage in control group had large red stain area, the red area was greatly reduced in KOA model and KOA+Len-NC group. The application of miR-27b-3p evidently increased the red area compared with that in KOA+Len-NC group (Fig.6B). The expression of BDNF、 p-TrkB and p-CREB in cartilage tissues were greatly inhibited in KOA and KOA+Len-NC groups compared with that in control group. The application of Len-miR-27b-3p evidently decreased their expression (Fig.6C).
Len-miR-27b-3p decreased Bax expression and promoted Bcl-2 expression that dysregulated in KOA and KOA+Len-NC group (Fig.6D、Fig.6E). The expression of TNF-α and IL-6 was greatly upregulated in KOA and KOA-Len-NC groups, and this upregulation was inhibited by Len-miR-27b-3p (Fig.6F、Fig.6G). These data showed miR-27b-3p protected by cartilage from degradation and inhibited inflammatory factors expression in vivo，to provide a good living environment for chondrocytes.

Discussion
The main pathological feature of KOA is degeneration and destruction of articular cartilage,because the cartilage cell lacks the nerve, the blood vessel and the lymph supply, as well as its own congenital differentiation and the migration ability is poor,the chondrocytes themselves have little ability to repair themselves,therefore, the current treatment of degenerative KOA cartilage is a major medical problem, how to repair the damaged articular cartilage and maintain its function is the treatment goal of osteoarthritis (23,24).
KOA is the leading disease that inducing disability in elderly people. However, there are no effective clinical treatment therapies to reverse the process of KOA (25,26). This could be the result of the still uncovered molecular mechanism. Thus, clarifying the mechanism and developing novel treatment target are of great importance.
In recent years, miRNAs have been proved to be involved in the development of KOA (11,26). Several miRNAs have been showed to regulate the pathogenesis of KOA (27)(28)(29). In this study, we confirmed miR-27b-3p was downregulated in IL-1β treated chondrocytes. Overexpressing miR-27b-3p inhibited IL-1β induced chondrocyte apoptosis and inflammatory factors secretion. Further, the target gene of miR-27b-3p was predicted to be BDNF. This prediction was proved by dual-luciferase reporter assay and BDNF mRNA and protein expression detection after miR-27b-3p overexpression. In in vivo study, miR-27b-3p overexpression inhibited cartilage degradation and inflammatory factors secretion. Thus, miR-27b-3p could exert positive role in KOA.
As chondrocyte is the only type cell in normal cartilage tissues, the apoptosis of chondrocyte in KOA greatly reduces the self-repair function of cartilage tissues.
Inhibiting chondrocyte apoptosis will reduce the degradation of extracellular matrix (ECM) and could be a novel treatment aspect (30). In this study, miR-27b-3p could reduce chondrocyte apoptosis via targeting BDNF. BDNF is upregulated in the joint tissue in KOA, and this upregulation could induce KOA related pain (31). As an extracellular cytokine, BDNF could induce the activation of an intracellular pathway, like TrkB/CREB (32). Activation of TrkB/CREB pathway by IL-1β could induce chondrocyte apoptosis. Inhibiting TrkB/CREB pathway exerts the beneficial role in KOA (33,34,35). In this study, the TrkB/CREB pathway was inhibited by miR-27b-3p mimics. This pathway was further activated by IL-1β+BDNF. Declarations corresponding author on reasonable request.

Authors' contributions
CLR design the research and revised the article; RJC and MW drafted the manuscript; LJW and YY collected the data and analyzed the data. XJL and HXL did the statistical analysis.

Ethical approval and consent to participate
This study was approved by the ethics committee of Yan'an University and in accordance with Tenth People's Hospital Affiliated to Tongji University.

Patient consent for publication
Not applicable.